Model Predictive Control using Prognosis and Health Monitoring of actuators

In order to keep a plant running until the next scheduled maintenance, a Prognosis and Health Monitoring (PHM) system may provide valuable information that can be used to distribute the control effort among the redundant actuators to avoid breakdown due to degradation. This paper proposes a Model Predictive Control (MPC) approach capable of distributing the control effort among actuators on the basis of PHM information, exploiting relations between control effort and actuator degradation. Constraints are then imposed to ensure that the accumulated actuator degradation will not reach a unsafe level at the end of M_H. A simulated tank level control system is used as a case study to illustrate the details of the proposed approach. The results show that the predictive controller distributes the control effort in a suitable manner to relieve the actuator with larger accumulated damage.